Rate This Document
Findability
Accuracy
Completeness
Readability

BoostMedia

BoostMedia accelerates media processing based on the Kunpeng architecture. It enables high-performance media solutions by leveraging deep vectorization and parallelization across audio/video encoding, decoding, and image processing scenarios.

Getting Started

  • What's New

    Provides the latest updates in the BoostMedia documentation.

Application Acceleration Features

  • x265

    x265 takes advantage of Scalable Vector Extension (SVE), a vectorized instruction set architecture, to improve the video encoding performance on Arm devices.

  • OpenCV

    This library optimizes common OpenCV functions with parallelization and vectorization to significantly boost image processing efficiency.

  • libavif

    This library uses methods such as vectorization, algorithm optimization, and workflow optimization to improve the image encoding performance of libavif.

  • libwebp

    This library uses methods such as vectorization, algorithm optimization, and workflow optimization to improve the image encoding performance of libwebp.

  • FFmpeg

    Vectorization is implemented for the sws_scale function and scaling algorithms (Bilinear, Bicubic, and Lanczos) within FFmpeg's libswscale library.

  • HW265Enc

    HW265Enc is an in-house video encoder based on the H.265/HEVC (High Efficiency Video Coding) standard. It encodes YUV pixel files to generate H.265/HEVC video bitstream files, and supports 8-bit color depth and 420p format.